context.vn
Menu

nl2 Repo

29 models evaluated

#ModelProviderTypeScore
1DeepSeek V4.1 FlashDeepSeekOpen weights65.4%
2DeepSeek V4 Pro 0813DeepSeekProprietary61.5%
3Ornith-1.5-397BOrnith AIOpen weights59.5%
4Hy4 previewTencent · Open weight58.9%
5GLM-5.3Z.AI · Open weight58%
6GLM-5.3-FlashZ.AI · Open weight56.3%
7Qwen3.8 MaxAlibaba · Open weight55.9%
8DeepSeek V4 Flash 0731DeepSeek · Closed54.2%
9dots3-note PreviewDots Studio · Open weight49.8%
10GLM-5.2Z.AI · Open weight48.9%
11Qwen3.8-Omni-FlashAlibaba · Closed48.9%
12Ornith-1.0-397BDeepReinforce AI · Open weight48.2%
13Qwen3.8-Flash-NextAlibaba · Open weight48.1%
14Qwen3.7 MaxAlibaba · Closed47.2%
15Seed 2.1 ProByteDance · Closed47%
16Ornith-1.5-35B-A3BOrnith AI · Open weight46.2%
17Seed 2.1 TurboByteDance · Closed43.7%
18Claude Opus 4.5Anthropic · Closed43.2%
19Qwen 3.6 Max (preview)Alibaba · Closed42.9%
20GLM-5.1Z.AI · Open weight42.7%
21Qwen3.8-27BAlibaba · Open weight42.3%
22MiniMax M3MiniMax · Open weight42.1%
23Qwen3.7 PlusAlibaba · Closed41.1%
24MiniMax M2.7MiniMax · Open weight39.8%
25Qwen3.6-27BAlibaba · Open weight36.2%
26Ornith-1.0-35BDeepReinforce AI · Open weight34.6%
27Ornith-1.5-9BOrnith AI · Open weight32.4%
28Qwen3.6-35B-A3BAlibaba · Open weight29.4%
29Ornith-1.0-9BDeepReinforce AI · Open weight27.2%